A novel aerated surface flow constructed wetland using exhaust gas from biological wastewater treatment: performance and mechanisms

In this study, a novel aerated surface flow constructed wetland (SFCW) using exhaust gas from biological wastewater treatment was investigated. Compared with un-aerated SFCW, the introduction of exhaust gas into SFCW significantly improved NH4+-N, TN and COD removal efficiencies by 68.30 ± 2.06%, 24.92 ± 1.13% and 73.92 ± 2.36%, respectively. The pollutants removal mechanism was related to the microbial abundance and the highest microbial abundance was observed in the SFCW with exhaust gas because of the introduction of exhaust gas from sequencing batch reactor (SBR), and thereby optimizing nitrogen transformation processes. Moreover, SFCW would significantly mitigate the risk of exhaust gas pollution. SFCW removed 20.00 ± 1.23%, 34.78 ± 1.39%, and 59.50 ± 2.33% of H2S, NH3 and N2O in the exhaust gas, respectively. And 31.32 ± 2.23% and 32.02 ± 2.86% of bacterial and fungal aerosols in exhaust gas were also removed through passing SFCW, respectively.


Introduction
Over the last few decades, rapid urbanization and economic growth has caused a series of severe environmental issues such as river pollution and water blooms in lakes, especially in developing countries. Considering the stringent discharge guidelines and standards for conventional wastewater treatment plants (WWTPs), WWTPs still face challenges in removing excess nutrients effectively from wastewater in an economical way, resulting in negative environmental consequences (Wu et al., 2016). On the one hand, WWTPs have not been constructed or fully operated due to their large capital investments and operating costs (aeration is the highest energy consumption period, accounting for 40-60% of the total plant operating costs) (Gu et al., 2008). On the other hand, constructed wetlands (CWs), which are regarded as a tertiary treatment process, have offered the greatest potential for secondary effluent treatment owing to their good efficiency, low costs and low maintenance (Vymazal, 2010).
Pollutant removal within CWs is a complex process that primarily includes substrate adsorption, plant absorption and microbial degradation. It is widely known that microbial nitrification and denitrification represent the major nitrogen removal mechanisms in CWs (Coban et al., 2015). According to the water level, CWs can be classified as surface flow (SF) CWs and subsurface flow (SSF) CWs. Compared to SSF wetlands, an SF wetland can better simulate natural systems, as the water flows over the bed surface and is filtered through a dense stand of aquatic plants, so SFCWs have often dominated in North America (Brix, 1994) reported that an effective increase in the ammonium nitrogen removal from 32.9-54.8% to 78.2-78.6% was achieved in aerated SFCWs. Even though artificial aeration is the most effective method of ensuring a sufficient oxygen supply, the corresponding operational cost greatly limits its popularity . Numerous studies have focused on improving the pollutant removal performance by using artificial aeration in SSFCWs (Hu et al., 2012a;Ong et al., 2010;Uggetti et al., 2016), but few studies have focused on using aeration to intensify the removal processes in SFCWs.
Hence, it is necessary to further optimize the oxygen supply strategy in SFCWs.
At present, the subject of focus in WWTPs is their pollutant removal efficiency and treatment cost but not the emission of exhaust gases. The exhaust gas produced by the aeration process often directly diffuses to the atmosphere, resulting in a nuisance to adjacent populations and a risk of serious environmental pollution. The exhaust gas can include bad odours (e.g., hydrogen sulphide (H 2 S) and ammonia release at the full scale. N 2 O has been listed as an important greenhouse gas that acts as the leading ozone depletion substance. Its 100-year global warming potential is 298 times higher than that of carbon dioxide (CO 2 ) (IPCC, 2013). In addition, microbial aerosols cannot be ignored, which is causing broad concerns all over the world (Brandi et al., 2000). In particular, pollution and the control of microbial aerosols from WWTPs have gradually become the focus of people's attention. Microbial aerosols are generated from the bursting bubbles produced by the aeration system.
The possible downwind movement of a microbial aerosol can increase the dispersion of airborne bacterial, viral and fungal species, which may represent a health risk for occupationally exposed personnel (Carducci et al., 1995). Srikanth et al. (2008) describes the impact of microbial aerosols on human health, and it is believed that the threshold limit value for microbial aerosols is very important for human health risk assessments. Thus, most WWTPs have become new pollution sources while also protecting the urban environment. However, the exhaust gas from biological wastewater treatment contains oxygen (O 2 ), CO 2 , volatile organic compounds (VOCs) and microorganisms, which are beneficial for improving plant growth and enhancing the microbial abundance in CWs. Until now, few people have realized that exhaust gas is a type of "available resource", and there have been no reports on using sewage treatment plant exhaust gas as a gas source for aerating SFCW.
As an initial attempt, the aim of this study was to design a novel aerated SFCW using exhaust gas from a biological wastewater treatment and investigating the treatment performance of the novel SFCW. For this purpose, three lab-scale SFCWs were operated under different conditions (un-aeration, intermittent aeration with air and intermittent aeration with exhaust gas). Simultaneously, the pollutant removal mechanisms in the SFCWs were investigated by measuring the microbe abundance. rhizomes per unit. After the transplantation, the CW systems were flooded using tap water and watered for two months until the sweet flag was well established. The effluent of SBR flow into a setting tank was then conveyed to the three SFCWs at a flow rate of 3 mL/min using a peristaltic pump. All the wetland systems were fed continuously with an HRT of 3 days. The pre-punched inflow and outflow tubes were positioned at the top of each tank at the same height. SFCWs B and C were intermittently aerated at an airflow rate of 0.012 m 3 /h for 2 h each cycle, and this rate was consistent with the SBR aeration time. SFCW A was operated without aeration.

Water sampling and analysis
Water samples were collected from the reactor influent tank and the effluent of each system (SBR and three SFCWs) every 3 days to analyse the transformation of and a pH meter (SG2, METTLER TOLEDO, Switzerland), respectively.

Plant sampling and analysis
At the end of June, July, August, September and October, plant leaves were harvested randomly from the different SFCWs and rinsed with distilled water to measure their chlorophyll contents. The leaf samples were cut into 1-2 cm square pieces and extracted for 24 h in the dark with 25 mL of 80% acetone. The total chlorophyll contents of the leaves were then determined using an ultraviolet spectrophotometer at 652 nm, as described by Bruinsma (1963). All the analyses were conducted in triplicate. The chlorophyll content was expressed based on the fresh weight (FW) (mg g -1 ).

The emission fluxes of N 2 O and CO 2
The N 2 O and carbon dioxide (CO 2 ) emission fluxes from the SBR and three SFCWs have been investigated in this study. When the experimental system tended to be stable, gas samples were taken in gas sampling bags (PV-500 ml; Delin, China)

The concentrations of odors gas and O 2
The odors gas (i.e., H 2 S and NH 3

Microbial aerosol
The bacterial and fungal aerosols from the SBR and SFCW C were collected using a six-stage Andersen sampler (Thermo-Andersen, Smyrna, GA, USA) which contained six Petri dishes filled with an appropriate agar medium. The Andersen sampler has six stages with different cutoff (D 50 ) sizes from higher to lower as follows:

Microbe sampling and analysis
At denitrifying bacteria (nirk, nirS and nosZ genes) were quantified by quantitative polymerase chain reaction (q-PCR).

Statistic analysis
All the statistical analyses were performed using SPSS statistical software package 11.0 (SPSS Inc., Chicago, USA). Two-sample t-tests were used to evaluate the significance of the differences among the means. The tables and figures show the results of the averaged data. In all the tests, the differences and correlations were considered statistically significant when P<0.05.

Nitrogen removal
After one month of operation, the effluent contaminants concentrations tended to be stable and   (Fig. 2b), which was ascribed to the oxygen content differences between the air (21.00 ± 0.13%) and exhaust gas (18.96 ± 0.16%). This result indicated that SFCW C could be beneficial in reaching full denitrification. Second, the CO 2 and volatile organic compounds (VOCs) in the exhaust gas could help contribute to the available carbon supply to promote the growth of nitrifying and denitrifying bacteria.
Third, in aerated SFCWs, aeration could cause a flow pattern change from laminar to turbulent so the mixing of pollutants in wastewater and microorganisms in the substrate would be extensive (Hu et al., 2012b). Moreover, the low TN removal efficiency in un-aerated CWs is primarily due to the poor nitrification caused by the DO deficit (Fig. 2b). inside the CWs to foster aerobic bio-degradation pathways of organics but could also stimulate anaerobic organics degradation. In addition, by comparison with SFCW B, a high COD removal rate was observed in SFCW C, which could be ascribed to the aeration with exhaust gas in the wetlands, promoting the growth and reproduction of heterotrophic bacteria.

Emergent plant monitoring
The chlorophyll content and plant height of the sweet flag in the three SFCWs were monitored every month during the experimental period, as shown in Table 1.

Purification of exhaust gas
Nitrogen removal has been improved, and simultaneously the exhaust gas has been purified. The change of microbial aerosol in exhaust gas after passing through SFCW was investigated, as shown in Fig. 4. A significant decrease in the particle concentrations of bacterial and fungal aerosols was observed. Results showed that bacterial and fungal aerosols in exhaust gas could be markedly reduced by 31.32 ± 2.23% and 32.02 ± 2.86% after passing through the SFCW, respectively. No Additionally, wetland plants also contribute to the removal of aerosols, which can prevent microbial aerosol diffusion.
Moreover, CWs also have good removal performance against H 2 S, NH 3 and N 2 O.
As shown in Fig. 5a, the removal efficiencies of H 2 S, NH 3 and N 2 O were 20.00 ± 1.23%, 34.78 ± 1.39% and 59.50 ± 2.33%, respectively. In this study, SFCW can be seen as a bioscrubber for purifying exhaust gas by water dissolution (leading role), times higher than that in SFCW B, because N 2 O in exhaust gas is only filtered through the liquid layer and then discharged directly into the atmosphere.

Microbial analysis
Microbial processes have been determined to be important pathways that contribute to nitrogen removal. The quantities of functional genes involved in biological nitrogen transformations, i.e., all bacteria (16S rRNA), AOB (amoA), NOB (nxrA), denitrifying bacteria (nirK+nirS) and nosZ genes are shown in Table 2.
There were remarkable differences in microbial abundance of the three SFCWs.
First, a large number of nitrifying bacteria (AOB, NOB) were detected in aerated SFCWs, while fewer AOB and NOB were detected in un-aerated SFCW, indicating that intermittent aeration was beneficial to promote the growth and reproduction of AOB and NOB, and the low DO in SFCW A seriously limited the growth of nitrifying bacteria. The results could explain the high removal of NH 4 + -N from aerated SFCWs (Fig. 2a). Second, the gene numbers of nirK and nirS exhibited an increase associated with the enhanced denitrifying activity that was responsible for eliminating the NO 3 --N concentration. The highest abundance of denitrifying bacteria was detected in SFCW C, followed by SFCW B and SFCW A. This result indicated that the aerated SFCWs only inhibited the activity of denitrifying bacteria rather than eliminating it.
Third, compared with SFCWs A and B, the abundance of nitrifying bacteria, denitrifying bacteria and total bacteria in SFCW C was greatly enhanced. The reason for this enhancement is probably that the microbe from the SBR was introduced to the CWs by aerating with exhaust gas.
In addition, to explain the microbial mechanism of N 2 O reduction, nosZ gene was conducted and the results are shown in Table 2. Reportedly, the nosZ gene that encodes the catalytic subunit of the N 2 OR has a plausible link to the N 2 O reduction activity, which was correlated with more bacteria being capable of reducing N 2 O to N 2 (Chourey et al., 2013). The highest nosZ copy density was observed in SFCW C, primarily because it had a higher quantity of denitrifying bacteria (Table 2) and better anoxic conditions (Fig. 2b), thus accelerating the expression of the nosZ gene.

Feasibility analysis in a real system
As an initial attempt, our present study designed a novel aerated surface flow constructed wetland using exhaust gas from biological wastewater treatment for mitigating WWTP exhaust gas emission and at the same time improving the performance of wetland. The collection and introduction of exhaust gas were crucial point to apply this system in a real system. On one hand, considering the rapid development of odour management and treatment for WWTPs, and various investigations on collection system of exhaust gases in WWTPs have been conducted Aerated constructed wetlands have been widely used in practical engineering applications. Thus, the results of present study might be easily applied in a real system without adjusting the existing technology.

Conclusions
A novel aerated SFCW using exhaust gas from biological wastewater treatment was designed and used to intensifying pollutants removal and increasing resource utilization. The introduction of exhaust gas significantly intensified the removals of organic and nitrogen. Simultaneously, the risk of exhaust gas pollution was effectively eliminated since SFCW (as a bioscrubber) could purify the exhaust gas ingredients, including H 2 S, NH 3 , N 2 O and microbial aerosols. Microbial analysis showed that the novel aerated SFCW could improve microbial abundance. The strategy of integrating intermittently aerated SFCW with exhaust gas has a promising future in purifying WWTP effluent and exhaust gas simultaneously.